Last week former Federal Reserve economist David Stockton’s report on Bank of England’s forecasting track record was published. City AM had this wrap-up (I didn’t read the report yet):
“INFLATION has been damaging British living standards and dragging down the economy – but the officials who are meant to keep a lid on prices didn’t do enough to help because their forecasts were too often wrong, according to a Bank of England report out today.
And even though the Bank was consistently worse at predicting changes in growth and inflation than other economists, it stuck with its flawed model, making excuses for its errors instead of trying to improve its forecasts.”
I would probably be less critical about that Bank of England’s forecasting abilities – or rather I know how hard it is to forecast anything, but I am not surprised to learn that Stockton find that BoE’s forecasts are biased. In fact I would be surprised if he had found that it was not biased. Central banks have strong incentives to do biased forecasts – and sometimes that might actually be what you want central banks to do. I for example find it very odd when central bank forecast that they will fail in achieving their policy objectives, but I also realize that central banks fail to hit their policy targets all the time.
David Stockton has 21 ideas to improve BoE’s forecasting abilities. Some of Stockton’s ideas are probably good, but I think that there is a more fundamental problem – and that is that central banks’ in-house forecasts very likely always will be biased. Therefore central banks should outsource forecasting – not because other institutions or companies (like banks!) necessarily are better at making forecasts than central banks, but because the forecasts of “outside” agents is likely to be much less biased than a in-house forecast.
One way would be to simply to outsource the forecasting to a private research company. Another possibility would be to base the forecast on a survey of professional forecasts – or even better as I have suggested numerous times that the central bank simply set-up a prediction market. In Britain that would be extremely easy – I don’t think there is a country in the world with so many bookmakers. The Bank of England could simply ask a company like Ladbrokes or a similar company to set-up betting markets for key macro economic variables – such as inflation and nominal GDP. It would be extremely cheap and the forecast created from such prediction market would likely be at least as good as what is presently produced by the otherwise clever staff at the BoE.
Related posts:
Yet another argument for prediction markets: “Reputation and Forecast Revisions: Evidence from the FOMC”
Benn & Ben – would prediction markets be of interest to you?
Prediction markets and government budget forecasts
Central banks should set up prediction markets
Scott’s prediction market
Ben maybe you should try “policy futures”?
What can Niskanan teach us about central bank bureaucrats?
Robin Hanson’s brilliant idea for central bank decision-making
Please fasten your seatbelt and try to beat the market
